Big Data Analytics: Its Architecture and Network

Model for Next Generation Technology

Md. Abu Salman, Student, Khulna University of Engineering and Technology , Khulna, Bangladesh

E-mail: abus02480@gmail.com

Big data is only getting bigger with the growth of artificial

Abstract — Big data is a new driver of the world economic
intelligence, social media and the Internet of Things with a
and social changes. The world’s data collection is
countless of sensors and devices. Data is measured in the
reaching a listing point for major technological changes
that can bring new ways in decision making, managing “3Vs” of variety, volume and velocity. There’s more of it
than ever before — often in real time. This severe flood of
our health, cities, finance, education and technology.
data is meaningless and unusable if it can’t be interrogated.
While the data complexities are increasing including
But the big data analytics model uses machine learning to
data’s volume, variety, velocity and veracity, the real
examine text, statistics and language to find previously
impact hinges on our ability to uncover the ‘value’ in the
data through Big Data Analytics technologies. Big Data unknowable insights. All data sources can be mined for
predictions and value [2].
Analytics poses a grand challenge on the design of highly
scalable algorithms and systems to integrate the data and Business applications range from customer personalization to
uncover large hidden values from datasets that are scheme detection using big data analytics. They also lead to
diverse, complex, and of a massive scale. Potential more efficient operations. Computing power and the ability
breakthroughs include new algorithms, methodologies, to automate are essential for big data and business analytics.
systems and applications in Big Data Analytics that The arrival of cloud computing has made this possible. Where
discover useful and hidden knowledge effectively. in next generation wireless advanced technology, big data
analytics can help to implement a theoritical as well as
Index Terms— Big Data Analytics, Algorithms, practical network concept about it.
Methodologies , Applications , Challenges.

II. ABOUT BIG DATA ANALYTICS

I. INTRODUCTION
Big data analytics is a complex process for analyzing a
My proposal on Big data analytics helps businesses, variety of large data sets or big data. This process reveals
organizations and upcoming technologies to make better information, such as hidden patterns, unknown correlations,
decisions by revealing information that would have otherwise and customer preferences, to ensure complete data
been hidden. management.

Meaningful visions about the trends, correlations and patterns Big data is collected from a wide variety of sources, such as
that exist within big data can be difficult to extract without videos, social networks, images, and sensors and needs to be
vast computing power. But the techniques and technologies processed like a analytical manner [1].
used in big data analytics make it possible to learn more from
large data sets. This includes data of any source, size and
structure.

The predictive models and statistical algorithms of data

visualization with big data are more advanced than basic
business intelligence queries. Answers are nearly instant
compared to traditional business intelligence methods. Figure 1: Big Data Analytics Concepts

1|P age
It is a form of advanced analytics, which involve complex analytics uses data mining, AI, and machine learning to
applications with elements such as analytical models, analyze current data and make predictions about the future.
statistical algorithms and advanced technologies powered by
analytics systems [3]. 4. Perscriptive Analytics: This type of analytics prescribes
the solution to a particular problem. This analytics works with
both descriptive and predictive analytics. Most of the time, it
III. METHODOLOGIES relies on AI and machine learning.

Data analysts, data scientists, predictive modelers,

IV. METHODOLOGIES FOR ADVANCED WIRELESS
statisticians and other analytics professionals collect, process, COMMUNICATION
clean and analyze growing volumes of structured transaction
data as well as other forms of data not used by others
programs [3]. In a service-driven next-generation network, a single
arrangement needs to effciently and exibly provide diversied
Here is the steps of the data preparation process: services such as enhanced mobile broadband, ultra-reliable
1. Data Collection and low-latency communications and massive machine type
2. Data Processing communications. It should also support coexistent accesses
3. Data Cleaning of multiple standards such the 5fth generation (5G), longterm
4. Data Analyzing evolution (LTE) and Wi-Fi. Also, it should coordinate a
heterogeneous network with different types of base stations
1. Data Collection: Data professionals collect data from a (BSs), for example, macro, micro, femto, pico BSs and
variety of different sources. Often, it is a mix of semi- diverse user devices as well as applications . The challenge to
structured and unstructured data. effciently operate a network capable of facilitating such
flexibility while satisfying the demands from diversied
2. Data Processing: Collected data must be organized, services is enormous for a network operator [4].
configure and partition the data properly for analytical
queries.

3. Data Cleaning: Data is cleansed for quality. That’s means

to remove errors or inconsistencies, such as duplications or
formatting mistakes, and organize and tidy up the data.

4. Data Analyzing: The collected, processed and cleaned

data is analyzed with the help of analytics software.

There are four types of big data analytics:

1. Descriptive Analytics
2. Diagnostic Analytics
3. Predictive Analytics Figure 2: Graphical illustration of the next-generation
4. Prescriptive Analytics communication system with some technological elements.

1. Descriptive Analytics: This summarizes past data into a

Operators have been optimizing their networks all along, but
form that people can easily read. This helps in creating
even today, the prevailing approach is to independently
reports, like a company’s revenue, profit, sales, and so on.
optimize single key performance indicators (KPIs), or an
Also, it helps in the tabulation of social media metrics.
element within the network independently , thus using a small
number of data sources. The network operators mostly
2. Diagnostic Analytics: This is done to understand what
depend on KPIs accumulated at different locations/parts of
caused a problem in the first place. Techniques like drill-
the network to make decisions employing various data
down, data mining, and data recovery are the examples.
analysis tools. Network monitoring and optimization are still
predominantly performed on old/recorded data, but this
3. Predictive Analytics: It looks into the historical and
dramatically restricts their capacity. The network operators,
present data to make predictions of the future. Predictive
in general, have/can have access to a vast amount of data from

2|P age
their networks and subscribers. With the appropriate
analytics, big data can convey broader intuitiveness and
understanding since it draws from multiple sources to reveal
previously unknown patterns and correlations . It benefits to
acquire a thorough knowledge of various unknown values
and delivers new measures in enhancing the performance
from different levels of wireless networks.

The value that analytics brings to optimization comes from

expanding the range of data sources and taking a
customercentric, quality of experience (QoE)-based approach
to optimizing end-to-end network performance [4].

Types of analytics for next generation technology:

Figure 3: Drivers from different directions that strengthen the case
There exists a succession of evolution in big data analytics, for the adoption of analytics in next-generation communication
systems.
starting from descriptive analytics to diagnostic analytics to
predictive analytics, and excelling towards prescriptive
analytics as shown in Fig. 3, out of which three (descriptive, The network operators have access to large amounts of data
predictive and prescriptive analytics) are dominant. The which can be categorized into two classes such as internal
network operators currently are in descriptive phase and use data and external data as shown in Fig. 4. The internal data
mainly the visualization tools to get insights on what has corresponded to data belonging to the network operators and
happened, the network performance, traffic profile, etc. The produced in the network, which is network related and
network operators can make use of the diagnostic analytics to subscriber related. The external data is collected from the
figure out the root-causes of the network anomalies and find third parties. Both the internal and external data can be further
out the faulty KPIs and network functions/elements. To get classified into two categories, which are structured data and
the diagnostic analytics, the analytics tool employs unstructured data. The structured is stored in a relational
techniques like drill-down, deep learning, data discovery, database, i.e., each field in the database has a name, and the
correlations, etc. Predictive analytics is an excellent tool for relationship between the areas are well-defined. On the other
making predictions. Note that it can never report or be precise hand, the unstructured data (for example, call center
about what will happen, however, predictive analytics can transcripts, messages, etc.) is not usually saved in a relational
only produce forecasting about what might occur, for database. Comprehensive coverage of the features and
example, future locations of the subscribers, future traffic sources of mobile big data can be found in [5] and [6].
pattern and network congestion, etc. Predictive analytics
deliver predictions events based on the real-time and archived
data by making use of various statistical techniques such as
machine learning, data mining, modeling as some analytical
process and game-theoretic analysis. Prescriptive analytics
goes steps ahead of just predicting the future events by
suggesting decision options for slicing (i.e., how to slice, how
many slices), virtualization, edge-computing, etc., along with
the implications of each decision option. Therefore, the
prescriptive analytics need a useful predictive model,
actionable data and a feedback system for tracking down the
results generated by the action taken. The decision options
(e.g., for network expansion, resource usage) are produced
considering the network operators preferences, system
constraints (backhaul, fronthaul, spectrum, transmission
power), etc. Prescriptive analytics can also suggest the finest
course of actions for any pre-defined target, for example, of
a particular KPI. Figure 4: Data sets and the sources of data available to
the network operators for big data analytics, machine
learning and artificial intelligence.

3|P age
 V. NETWORK MODEL FOR NEXT GENERATION VII. APPLICATIONS OF BIG DATA
TECHNOLOGY
Big data has found many applications in various fields
today. The major fields where big data is being used are as
follows.

1. Government

Big data analytics has proven to be very useful in the

government sector. Big data analysis played a large role in
Barack Obama’s successful 2012 re-election campaign. Also
most recently, big data analysis was majorly responsible for
the BJP and its allies to win a highly successful Indian
General Election 2014. The Indian Government utilizes
numerous techniques to ascertain how the Indian electorate is
responding to government action, as well as ideas for policy
Figure 5: Network Model of Big Data Analytics in next augmentation [8].
generation technology
2. Social Media Analytics
In the network model of next generation technology, data The advent of social media has led to an outburst of big data.
will be analyzied by four steps process. First is data
Various solutions have been built in order to analyze social
acquisition. In the step data will be generated as many forms
media activity like IBM’s Cognos Consumer Insights, a point
and transmitted to the next step. Second step is data
solution running on IBM’s Big Insights Big Data platform,
processing and storage. Data will be processed and stored in
can make sense of the chatter. Social media can provide
the step by the help of different types of BSs ( Base Stations).
valuable real-time insights into how the market is responding
In third step, which is data transportation, data will form a
to products and campaigns. With the help of these insights,
network and transported to the data center. Where in last and
the companies can adjust their pricing, promotion, and
final step, data analytics will occur in data center.
campaign placements accordingly. Before utilizing the big
data there needs to be some preprocessing to be done on the
The proposed big data analytics technology for wireless
big data in order to derive some intelligent and valuable
networks will develop a comprehensive theoretical
results. Thus to know the consumer mindset the application
understanding, practical system design guidelines, and
of intelligent decisions derived from big data is necessary [8].
demonstration of a big data aware intelligent demand-
attentive wireless cloud network that adapts itself originally 3. Technology
and dynamically to the user demands and service
requirements in order to improve the delivery of multimedia The technological applications of big data comprise of the
content and the end-user Quality of Experience (QoE) [7]. following companies which deal with huge amounts of data
every day and put them to use for business decisions as well.
For example, eBay.com uses two data warehouses at 7.5
VI. ARCHITECTURE MODEL OF BIG DATA petabytes and 40PB as well as a 40PB Hadoop cluster for
search, consumer recommendations, and merchandising.
Inside eBay’s 90PB data warehouse. Amazon.com handles
millions of back-end operations every day, as well as queries
from more than half a million third-party sellers. The core
technology that keeps Amazon running is Linux-based and as
of 2005, they had the world’s three largest Linux databases,
with capacities of 7.8 TB, 18.5 TB, and 24.7 TB. Facebook
handles 50 billion photos from its user base. Windermere
Real Estate uses anonymous GPS signals from nearly 100
million drivers to help new home buyers determine their
typical drive times to and from work throughout various
Figure 6: Architecture of Big Data Analytics times of the day [8].

4|P age
4. Fraud detection the industry have improved their ability to work with such
data, even though the files are enormous and often have
For businesses whose operations involve any type of claims different database structures and technical characteristics [8].
or transaction processing, fraud detection is one of the most
compelling Big Data application examples. Historically, Big data in healthcare (In Broad scene):
fraud detection on the fly has proven an elusive goal. In most
cases, fraud is discovered long after the fact, at which point The Internet of Healthcare Things (IoHT) is progressing
the damage has been done and all that’s left is to minimize toward a smart, intelligent, reliable, and automated structure
the harm and adjust policies to prevent it from happening with the rapid adoption of various evolving techniques (such
again. Big Data platforms that can analyze claims and as artificial intelligence, advanced machine learning, big data
transactions in real time, identifying large-scale patterns analysis, pattern recognition) and systems (such as the
across many transactions or detecting anomalous behavior Internet of Things (IoT) and robotics) .The upcoming
from an individual user, can change the fraud detection game Intelligent Internet Healthcare Things (IIoHT) demands
[8]. secure, ultra-reliable, low latency, ultra-high data rate
communication among a massive number of IoHT nodes.
5. Telecommunication

Now a day’s big data is used in different fields. In telecom

also it plays a very good role. Operators face an uphill
challenge when they need to deliver new, compelling,
revenue-generating services without overloading their
networks and keeping their running costs under control. The
market demands new set of data management and analysis
capabilities that can help service providers make accurate
decisions by taking into account customer, network context
and other critical aspects of their businesses. Most of these
decisions must be made in real time, placing additional
pressure on the operators. Real-time predictive analytics can
help leverage the data that resides in their multitude systems,
make it immediately accessible and help correlate that data to
generate insight that can help them drive their business Figure 7: A 6G communication scenario for IIoHT
forward [8].
The upcoming 6G wireless system will deliver short-range
6. Healthcare wireless connectivity with terabit speed (0.1–10 THz), which
is ideally suited for transmitting large-scale medical data to
Traditionally, the healthcare industry has lagged behind other
the IIoHT arrangement [9].
industries in the use of big data, part of the problem stems
from resistance to change providers are accustomed to
VIII. CHALLENGING ISSUES OF BIG DATA
making treatment decisions independently, using their own ANALYTICS AND HOW TO SOLVE THEM
clinical judgment, rather than relying on protocols based on
big data. Other obstacles are more structural in nature. This is
one of the best place to set an example for Big Data Big data analytics have some challenging things that are as
Application. Even within a single hospital, payer, or follows.
pharmaceutical company, important information often
remains soloed within one group or department because 1. Lack of proper understanding of Big Data
organizations lack procedures for integrating data and
communicating findings. Companies fail in their Big Data enterprises due to
insufficient understanding. Employees may not know what
Health care stakeholders now have access to promising new data is, its storage, processing, importance, and sources. Data
threads of knowledge. This information is a form of “big professionals may know what is going on, but others may not
data,” so called not only for its sheer volume but for its have a clear picture.
complexity, diversity, and timelines. Pharmaceutical industry
experts, payers, and providers are now beginning to analyze For example, if employees do not understand the importance
big data to obtain insights. Recent technologic advances in of data storage, they might not keep the backup of sensitive

5|P age
data. They might not use databases properly for storage. As a Solution:
result, when this important data is required, it cannot be
retrieved easily. The best way to go about it is to seek professional help. You
can either hire experienced professionals who know much
Solution: more about these tools. Another way is to go for Big Data
consulting. Here, consultants will give a recommendation of
Big Data workshops and seminars must be held at companies the best tools, based on your company’s scenario. Based on
for everyone. Basic training programs must be arranged for their advice, you can work out a strategy and then select the
all the employees who are handling data regularly and are a best tool for you.
part of the Big Data projects. A basic understanding of data
concepts must be inculcated by all levels of the organization. 4. Lack of data professionals

2. Data growth issues To run these modern technologies and Big Data tools,
companies need skilled data professionals. These
One of the most pressing challenges of Big Data is storing all professionals will include data scientists, data analysts and
these huge sets of data properly. The amount of data being data engineers who are experienced in working with the tools
stored in data centers and databases of companies is and making sense out of huge data sets.
increasing rapidly. As these data sets grow exponentially with
time, it gets extremely difficult to handle. Most of the data is Companies face a problem of lack of Big Data professionals.
unstructured and comes from documents, videos, audios, text This is because data handling tools have evolved rapidly, but
files and other sources. This means that you cannot find them in most cases, the professionals have not. Actionable steps
in databases. need to be taken in order to bridge this gap.

Solution: Solution:

In order to handle these large data sets, companies are opting Companies are investing more money in the recruitment of
for modern techniques, such skilled professionals. They also have to offer training
as compression, tiring, and deduplication. Compression is programs to the existing staff to get the most out of them.
used for reducing the number of bits in the data, thus reducing Another important step taken by organizations is the purchase
its overall size. Deduplication is the process of removing of data analytics solutions that are powered by artificial
duplicate and unwanted data from a data set. Data tiring intelligence/machine learning. These tools can be run by
allows companies to store data in different storage tiers. It professionals who are not data science experts but have basic
ensures that the data is residing in the most appropriate knowledge. This step helps companies to save a lot of money
storage space. Data tiers can be public cloud, private cloud, for recruitment.
and flash storage, depending on the data size and importance.
Companies are also opting for Big Data tools, such 5. Securing data
as Hadoop, NoSQL and other technologies.
Securing these huge sets of data is one of the
3. Confusion while Big Data tool selection daunting challenges of Big Data. Often companies are so
busy in understanding, storing and analyzing their data sets
Companies often get confused while selecting the best tool that they push data security for later stages. But, this is not a
for Big Data analysis and storage. Is HBase or Cassandra the smart move as unprotected data repositories can become
best technology for data storage? Is Hadoop breeding grounds for malicious hackers.
MapReduce good enough or will Spark be a better option for
data analytics and storage? Companies can lose up to $3.7 million for a stolen record or
a data breach [10].
These questions bother companies and sometimes they are
unable to find the answers. They end up making poor Solution:
decisions and selecting an inappropriate technology. As a
result, money, time, efforts and work hours are wasted. Companies are recruiting more cybersecurity professionals to
protect their data.

6|P age
 IX. CONCLUSION

Here I consider a next-generation wireless network model, [5] X. Cheng, L. Fang, L. Yang, and S. Cui, ``Mobile big
data: The fuel for data-driven wireless,'' IEEE Internet Things
where the network operators employs advanced data J., vol. 4, no. 5, pp. 1489_1516, Oct. 2017.
analytics, ML and AI for efficient operation, control, and
optimization. I present the main scheme of big data analytics [6] X. Cheng, L. Fang, X. Hong, and L. Yang, ``Exploiting
adoption and discuss how ML, AI and computational mobile big data: Sources, features, and applications,'' IEEE
intelligence play their important roles in data analytics for Netw., vol. 31, no. 1, pp. 72_79, Jan./Feb. 2017.
next-generation wireless networks. I present a set of network
[7] T. Ratnarajah, “Big Data Aware High Capacity Wireless
design and optimization schemes with respect to data
Network Architecture Using Caching and Machine leaning”,
analytics. Finally, I discuss the paybacks and challenges that Accessed on: April 19, 2021. [Online].
the network operators encounter in adopting big data Available:http://www.profratnarajah.org/big-data-analytics-
analytics, ML, and AI in next-generation wireless networks. and machine-leaning-for-wireless communications.html

X. ACKNOWLEDGEMENT [8] Deepak Kumar Sahu , All about Big Data Applications,
Data Analytics, Accessed on: April 19, 2021. [Online].
Available: https://www.digitalvidya.com/blog/big-data
This report was prepared by me and the research was applications/

directed under a course of “Advanced Wireless [9] Kaiser M.S. et al. (2021) 6G Access Network for
Communication” for M.Sc. Eng. degree in Electrical and Intelligent Internet of Healthcare Things: Opportunity,
Electronic Engineering, conducted by Prof. Dr. Mostafa Challenges, and Research Directions. In: Kaiser M.S.,
Zaman Chowdhury. Bandyopadhyay A., Mahmud M., Ray K. (eds) Proceedings
of International Conference on Trends in Computational and
So I would want to give thanks to my honorable sir Prof. Dr. Cognitive Engineering. Advances in Intelligent Systems and
Computing, vol 1309. Springer, Singapore.
Mostafa Zaman Chowdhury for giving me the opportunity to
prepare a report on my research. [10] Rohit Sharma, Top 6 Major Challenges of Big Data &
Simple Solutions To Solve Them, Accessed on: April 19, 2021.
XI. REFERENCES [Online]. Available: https://www.upgrad.com/blog/major-
challenges-of-big-data/
[1] M. Z. Chowdhury, Md. Shahjalal, S. Ahmed, Y.M Jang,
“6G Wireless Communication Systems: Applications,
Requirements, Technologies, Challenges, and Research
Directions”, IEEE Open Journal of the communications
Society, Vol. 1, 2020.

[2] Big Data Analytics - A Complete Introduction, What is

Big Data Analytics? , Accessed on: April 19, 2021. [Online].
Available:https://www.omnisci.com/learn/big-data-analytics

[3] Wesley Chai, “big data analytics”, Accessed on:

April 19,2021.[Online].Available:https://searchbusinessanal
ytics.techtarget.com/definition/big-data-analytics

[4] M. G. Kibria, K. NGUYEN, G.P Vallardi, O. Zhao, K.

Ishizu, F. Kojima “Big Data Analytics, Machine Learning,
and Artificial Intelligence in Next-Generation Wireless
Networks”, IEEE Access, Wireless Systems Laboratory,
Wireless Networks Research Center, National Institute of
Information and Communications Technology, Yokosuka
239-0847, Japan.

7|P age